Radical, Anion, Anion-Radical Reactions with Organic Halides

21 Radical, Anion, Anion-Radical Reactions with Organic

Downloaded by UNIV OF CALIFORNIA SAN DIEGO on February 5, 2016 | http://pubs.acs.org Publication Date: June 1, 1978 | doi: 10.1021/bk-1978-0069.ch021

Halides SHELTON BANK and JANET FROST BANK Department of Chemistry, State University of New York at Albany, Albany, NY 12222

Interesting similarities in the behavior of the three reactive species, having one or two electrons, led us to study their reactions with organic halides from the point of view that comparisons of product and kinetic data could lead to correlations, and of equal importance the lack thereof, of the various mechanisms. Aromatic radical anions have been studied in depth beginning in 1867 with Bertholet's discovery (1), and continuing with Schlenk and Bergmann's pioneering work (2) around the turn of the century. Although radical anions exhibit complex chemical behavior, their a b i l i t y to function as reducing agents by electron transfer processes is reasonably well understood. In this regard, reduction of organic halides was selected as a model for election transfer reactions of radical anions. The products of the reaction of radical anions, for example sodium naphthalene, with bromides and chlorides are the hydrocarbon, the olefin and alkylated dihydronaphthalene. Reaction with iodides gives hydrocarbon dimer in addition. The mechanism as shown is a result of numerous product and kinetic studies (3, 4, 5) Two steps that merit further investigation are the

©0-8412-0421-7/78/47-069-343$05.00/0

In Organic Free Radicals; Pryor, William A.; ACS Symposium Series; American Chemical Society: Washington, DC, 1978.

ORGANIC FREE RADICALS

344

(3)


+ X

i n i t i a l electron transfer (eq. 1) and the coupling reaction (eq.2). Two p o s s i b i l i t i e s for the f i r s t step are that the species, RX , is a transition state for a single-step bond dissociation, and that the species is an intermediate organic halide radical anion. The f i r s t possibility is generally accepted, however, extensive kinetic data (4C) for a series of organic radical anions and halides indicate a change in the nature of the transition state in going from a highly exothermic to a less exothermic reaction. These changes could result from differences in timing along the reaction path in a one-step bond cleavage, or from a superposition of the two proposed mechanisms. In an attempt to resolve this dilemma, the reaction rates were measured for a series of organic bromides with sodium anthracene (Table I) and correlated with two model systems. The model for the f i r s t p o s s i b i l i t y , a one-step process with bond dissociation, is the t r i - n - b u t y l t i n hydride reaction with these same halides { § ) . Correlation of the reaction rates would indicate that the transition state for anthracene radical anion reduction is similar to the transition state for radical formation by tri-jQ-butyltin radical. On the other hand, the model for the second p o s s i b i l i t y , the organic halide reduction potent i a l , is a measure of organic halide radical anion formation (7J, T

Table I. Second-Order Rate Constants for Reaction of Sodium Anthracene with Alkvl Bromides in THF at 0°C. LoqKfrel.i kfrelj k, M- secAlky! Bromide 0.0 1.0 242 jl-Butyl 0.59 3.9 938 Sêç-Butyl 1.0 2402c 9.9 tent-Butyl -2.82 0.0015 Phenyl 25. pe 1,40 Benzyl (a) Measured with stopped-flow apparatus at 775 nm (b) The rate was unaffected by the addition of tert -butanol in excess, (c) tert- Butanol was added to prevent build-up of strongly absorbing anion, (d) From ref 4c at 20°C. (e) Estimated from competition experiments (8, 9). L

b c

d

The relationship in Figure 1 between the known radical reaction and the radical anion reaction is monotonie and there-


21.

BANK AND BANK

Anion

and

Radical

Reactions

345


• Phenyl

n-Butyl

2-Butyl

·

t-Butyl

Benzyl

^ 0

J

7!7

Log krel Bu^Sn* Figure

1.

The relationship dium anthracene

between the logs of the relative rates of reaction and tri-n-butyltin hydride with select bromides


of so-


346

ORGANIC F R E E RADICALS

fore similar reaction factors are likely to be involved. This definitive correlation compared to the decided lack of correlation in Figure 2 indicates that for sodium anthracene the formation of R X * is not involved to any great extent, but rather the transition state resembles that for a radical reaction. In contrast, for reaction of organic halides with metallic magnesium, Whitesides {7J found a poor correlation with the rates of t r i - n - b u t y l t i n hydride, and a reasonable correlation, especially with primary bromides, with reduction potentials, suggesting the formation of RX7 . Sodium naphthalene reductions also correlate with reduction potentials for primary halides (4b). Moreover, reaction rates are faster in solvents favoring loose ion pairs over tight, further evidence for an early transition state involving election transfer and l i t t l e bond dissociation. (4c) Therefore, one can conclude that for the less exothermic reduction by sodium anthracene,eq. 1 is a one-step radical formation. For radical anions of higher reduction potential, the possibility for the species, R X T , must s t i l l be considered. Turning now to equation 2 in the mechanism, the genesis of the alkylated aromatic raises the question of possible contributions from displacement reactions, and more generally the question of possible stereochemical integrity. Steps 5 and 6 were considered

to contribute foremost to the mechanism (10) before experiments by Sargent (5b) and Garst (3c) showed l i t t l e dependence either on the structure or on the halogen of the alkyl halide in the ratio of alkylated to reduced products. Data from experiments with chiral substrates (11) however, suggests that an SN2 process might contribute to some extent. An important prediction of such a path is of course inversion of configuration at the reaction s i t e . This is moreover a stringent test for the timing of steps 1 and 2A, for i f there is a stereochemical preference for step 2A and i f reaction is faster than racemization of the r a d i c a l , then stereochemical integrity can be preserved. Bridgehead tertiary halides were chosen as the experimental model because rear-side attack is not possible, and therefore, i f displacement is involved, there would be a decrease in the amount of alkylated products. The results in Table II are consistent


BANK AND BANK

Anion

and

Radical

Reactions

Benzyl


1.0-1

•

t-Butyl

>n-Butyl

r-

Φ

l.OH

SCD Ο

2.0-

Phenyl 3.0·

i X

2.4 -Ε (V vs SCE)

Figure 2. The relationship between the log of the relative rates of reaction of sodium anthracene with select bromides and their reduction potentials

American Chemical Society Library 1155 16th S t N. W.

In Organic Free Radicals; Pryor, William A.; Washington, D. 6. Society: 20Θ36Washington, DC, 1978. ACS Symposium Series; American Chemical

348


Table II.

Reaction of Bridgehead Bromides with Sodium

Napntnaiene Bromide 1 -Adamant^/I 1-Bicyclo "2.2.2.]octyl 4-Bicyclo "2.2.2.]-l-azooctyl Downloaded by UNIV OF CALIFORNIA SAN DIEGO on February 5, 2016 | http://pubs.acs.org Publication Date: June 1, 1978 | doi: 10.1021/bk-1978-0069.ch021

T_

(b) (c)

T i l l -

» i

R-Naph-Η*· ·

R-H% 41 » 42 50

1

590

(58) (50)

b

b

' ΐ Λ Ο Λ

Analyses were by vpc on a U.C. W 98 column. Determined by difference from the yields of aliphatic products. Isolated by liquid chromatography and characterized by mass spectroscopy and proton m m . r.

(d)

with yields from primary halides. (4b) They agree as well with experiments by Sargent (5b) who found 61% alkylation from reac tion with tert-pentyl iodide. In a l l cases i t appears that the alkylation reaction is insensitive to stereochemistry which in turn is inconsistent with an SN mechanism. The insensitivity further indicates that the r a d i c a l , which is clearly implicated, has sufficient time to equilibrate before reaction. These results additionally provide estimated rate limits for reactions bearing on stereochemical integrity. From competition experi ments, the rate constant for the coupling step was determined to be ~1χ10 M" sec (12). Accordingly, loss of stereochemical integrity must be faster to account for the commonality of results with the acyclic and bridgehead bromides. A parallel scheme for electron transfer processes with anions (eq. 8 and 9) is suggested and this precludes any stereochemical integrity unless there is some special feature about the coupling step. The tenfold increase in rate (Table I) in going from a primary to a tertiary bromide is in direct contrast to the expected structural effect for a displacement reaction. This supports further the conclusion that alkylation products arise neither by an SN2-like process, nor by any process with strong stereochemical requirements. A potential, common surface for the chemistry of anions and radicals involves organometallic compounds and organic halides. In addition to the usual two-election displacement (eq. 7), a stepwise scheme (eq. 8 and 9) can explain the products of reac t i o n . The electron transfer step (eq£) has been implicated by 2

9

1

An" + RX —> An-R + X" An" + RX An* + R + X" An* + R —> An-R e

#

(7) (8) (9)

the observation of radicals using ESR (13) and CIDNP (14) tech niques. These observations prompted a search for the quantita tive contributions of radicals and their role in the transition state for anion reactions.



21.

BANK AND BANK

Anion

and

Radical

349

Reactions

To provide the quantitative data we have focused attention on the kinetics of the reaction of select anions with halides. This approach concentrates on the rate determining step and the factors that affect i t . Two kinds of anionic species have been investigated, 9-alkyl-10-1ithio-9,10-dihydroanthracenyl anions and l i t h i o di-and tri-pehnylmethyl anions. The stereochemistry of the alkylanthracenyl anion reactions with halides has in fact fascinating results. In the main, reaction with primary halides leads to a predominance of the cis product, while reaction of the secondary halides leads to the trans (15). We have used kinetic measurements to sort out the factors influencing the transition states for cis and trans products.

+ RBr

-h Li Br

(lu)

Table III. Second-Order Rate Constants for Reaction of 9-Alkyl1O-Lithio-9JO-Dihydroanthracene with Alkvl Bromides α Bromide oZKTUTïi UT7*nw YD c / T T » a n c νk(cis) (rî cir Ts/trans k(trans) 26/3 n-Hexyι H 76/24 1124 346 1473 Et 237 337 59/41 iPr 574 280 31 10/90 312 tBu 62 i-Propyl H 30 25/75 10 Et 40 5 32 13/87 iPr 37 0.7 35 2/98 36 tBu (a) Measured with stopped-flow apparatus described previously, in THF at 20° at 400 nm. (b) Absolute rate constants were corrected for the elimination reaction (kl). The opposite is true for naphthalene where the radical anion is faster than the anion in reaction with primary chloride. The difference arises from disparate reactivity of the respective radical anions and shows the danger in extrapolating from one radical anion-anion pair to another. Another point of interest is the much greater reactivity of naphthalene radical anion vs. anthracene, which suggests alternate mechanisms for the electron transfer step,again raising the possibility for the existence of RX? for the naphthalene system. The comparative reactivity order for Anth" ys.. Anth"" with chlorides and bromides is contrary to what is expected on the basis of their respective reduction potentials (17). This rate ratio changes by an order of magnitude in going from chlorides to bromides to iodides. For the iodides the rates of the two reactions are comparable but this surprising similarity may be fortuitous and may not in fact mean a similarity in mechanisms. Further discussion of this point appears in a subsequent section. Focusing last on the rate ratios of d i - and t r i phenylmethyl anions, we note that in contrast to the reduction potential order (E*53(r = 1.3v(18)>E^ Cir = 1.1 ν (19.)) but in concert with the basicity order (pka Φ^ΟΗ" = 33.1 > pka Φ30- = 31.5) (20), the diphenylmethyl anion is more reactive. This rate difference of about two powers of ten is ~3kcal/mole in AGt for primary and secondary bromides and decreases to 1.7 kcal/mole in AGt for primary and secondary iodides. It appears as i f factors affect ing basicity contribute more than factors affecting reduction potential for both primary and secondary systems. The proposed comparisons of r a d i c a l , radical anion and anion reactions are now described by two reactivity grids for organic halide reductions. Table V records patterns for a wide variety of reagents. There are definite trends, however for the chlo rides, Figure 3 shows significant overlapping of reactivities of the three species. In particular, at log reactivity ratio of -2.7, there is an overlapping of a l l three reagents. The kinetic data as such is not l i k e l y to provide clear distinctions of mechanistic paths. On the other hand reactivity ratios for the iodides offer a useful diagnostic for mechanistic distinctions. This is based on two considerations. F i r s t , there is no overlap of the three species. The f u l l spread of SN2 values for both dipolar aprotic solvents as well as protic solvents is quite distinct from the radical values, which in turn are distinct from radical anion valves. Second, the differences are s u f f i c i e n t l y large to be useful. At the midpoints of the radical anion, radical and SN2 ranges, the iodides would be 282, 68 and 6 times more reactive than the corresponding bromides. These values should lend confidence to predictions. The second reactivity grid describes the structural effects 2




Figure 3. The relationship between the log of relative rates of reaction of 9-alkyl-10-lithio-9,10dihydroanthracenyl anions with hexyl bromide, and the Τ aft steric constants for the alkyl groups


21.

BANK AND BANK

Anion

and

Radical

353

Reactions


of the halide. The rate ratio of primary to secondary halides in Table VI f a l l into a pattern such that radical and radical anion reactivities are similar and at midpoint the secondary halide is three times more reactive than the primary. Displacement reactivity ratios are distinct and the primary is 10-15 times more reactive than the secondary. Table VI. The Ratios of Reaction Rates for Primary vs. Secondary Halides with Various Reducing Reagents ··

— -

Rif;

Bu Sn'H + RBr

0.33

C r ( e n ) H + RBr

0.11

(21.)

Phenyl* + RI

0.57

An" + RBr

0.26

(27) (4e)

3

2

average

15.5

(24,

Φ 0Η" + RBR

14.2

(26)

Φ 0Η" + RI

10.1

(26)

SN

2

2

2

9-RAnH" + RBr

9.0

(trans)

25)

(26)

We now attempt to u t i l i z e these diagnostics for our study of organometallic reagents and the reaction scheme in equations 7, 8 and 9. In table V, the reactivity ratios for chlorides i s , as expected in the overlapping region of r a d i c a l , radical anion and SNg mechanisms. The iodide ratio is d i s t i n c t , however, and although the reactivities are high and near those for radicals, the patterns for structural changes correspond to displacement reactions. In this regard, for reaction of alkyl dihydroanthra cenyl anions to give trans product, the similarities of the rate ratios with those of other anions mitigates against the mecha n i s t i c change. The predominate mechanistic pathway for the reaction of these anions and halides is consistent with a two-electron process (eq. 7), however this kind of kinetic evidence cannot exclude contributions of ~10% from an alternative scheme and clearly any contributions below 1% are undetectable. To assess small contri butions to any certainty will probably require structural changes that augment those contributions. In conclusion, do the similarities and differences in the reactivity patterns for these reactive species with organic halides indicate a basic relationship or are these similarities fortuitous and does nature provide distinct pathways of comparable energy. Of the three mechanistic p o s s i b i l i t i e s considered for radical anions, electron transfer, r a d i c a l - l i k e and nucleophic substitution, electron transfer is indicated for highly exothermic reactions, and a r a d i c a l - l i k e process is l i k e l y for the less




354

Figure

4.

Correlations

of anion, radical, and radical anion reactions halides and the halide effect

with


organic

BANK AND BANK

21.

Anion

and

Radical

355

Reactions

Table V. The Log of the Ratio of Second-Order Rate Constants for Chlorides and Iodides vs. Bromides with Various Reducing Reagents. Reagent

RI

Reference

-2.82

1.80

(II)

-2.15

1.85

Py* + Benzyl

-3.47

Naph

-2.58

2.50

(4d)

Pyrene"

-2.96

2.27

(4d)

Anth

-3.71

2.57

(4d)

Bu SnH 3

Cr(en) Downloaded by UNIV OF CALIFORNIA SAN DIEGO on February 5, 2016 | http://pubs.acs.org Publication Date: June 1, 1978 | doi: 10.1021/bk-1978-0069.ch021

RC1

2

7

7

(6) (21_) (22)

Py* + N0 -Benzyl

-3.36

SN Typical

-1.7

0.7

(24)

φε" in EtOH

-2.1

0.54

(25)

2

2

(23)

EtO" in EtOH

-2.62

0.28

(25)

N ~ in DMF

-2.66

0.83

(25)

N ~ in MeOH

-2.13

1.05

(25)

SCN" in DMF

-2.67

0.84

(25)

SON" in MeOH

-2.31

0.48

(25)

Φ^Η" in THF

-2.44

1.35

(26)

HAnH-in THF

-2.43

0.91

(26)

3

3




356

exothermic. Except for the fact that energetics of reaction sometimes are comparable for radical anion and displacement reactions (9), there is no compelling evidence that mechanistic similarities apply. In this regard a better concept of radical anions is as radicals with high reduction potentials rather than as anions. Moreover, the relatively weak basicity and reactivity towards proton donors does not f i t well with connotations of anions (28). For the case of anions, r a d i c a l - l i k e or pathways involving single electron transfer are not in evidence from the work described here. The products and reactivity patterns are ex pressed adequately by two-electron processes. A c r i t i c a l difference between the one and two-electron process is the con tribution of bond making to the l a t t e r . Apparently this provides a substantial energy lowering that makes this path dominate in the systems studied in spite of the fact that electron transfer is possible. Conceivably this situation will be reversed for these kinds of anions when bond making is rendered less important, or when bond breaking more important, as for example via steric effects. Acknowledgments. We are appreciative of the assistance of coworkers and collaborators whose names appear in the references. The financial assistance of NATO Grant No. RG 1069 is gratefully acknowledged. Literature Cited (1). (2). (3).

(4).

(5).

Berthelot, Μ., Ann. Chem. (Paris) (1867), 12, 155. Schlenk, W. and Bergmann, E., Annalen, (1928), 464, 1, and references cited therein. (a). Garst, J . F . , Accounts of Chemical Research (1971), 4, 400. (b). Garst, J . P . , Barbas, J . T., and Barton, F. E., J . Am. Chem. Soc., (1968), 90, 7159. (c). Garst, J . F . , Roberts, R. D . , and Abels, Β. N . , J . Am. Chem. S o c ., (1975), 97, 4926 and references cited therein. (a). Sargent, G. D . , Cron, J . N. and Bank, S . , J. Am. Chem. Soc., (1966), 88, 5363. (b). Bank, S. and Bank, J. F . , Tetrahedron Lethers, (1969), 5433. (c). Bank, S. and Juckett, D. Α . , J. Am. Chem. Soc., (1975), 97, 567. (d). Bank, S. and Juckett, D. Α . , J. Am. Chem. Soc., (1976), 98, 7742. (a). Sargent, G. D . , and Browne, M. W., J. Am Chem. Soc., (1967), 89, 2788. (b). Sargent, G. D . , and Lux, G. Α . , J. Am. Chem. Soc.,


21.

BANK AND BANK

Anion

and Radical

Reactions

357

(1968), 90, 7160. Sargent, G. D . , Tetrahedron Letters, (1971), 3279 and references therein. (6). (a). Kuivila, H. G., Accounts of Chemical Research (1968), 1, 299. (b). Kuivila, H. G. Menapace, L. W., and Warner, C. R., J . Am. Chem. Soc., (1962), 84, 3854. (c). Menapace, L. W. and Kuivila, H. G., J. Am. Chem. Soc., (1964), 86, 3047. (7). Rogers, R. J., Mitchell, H. L., Fujiuara, V., and Whitesides, G. M . , J. Org. Chem., (1974), 39, 857. (8). (a). Lee, Y . , Ph.D. Thesis, SUNY at Albany, Sept., 1974. (b). Lee, Y. and Closson, W. D . , Tetrahedron Letters, (1974), 1395. (9). Bank, S . , and Thomas, S. P . , J. Org. Chem., (1977), 42, 2858. (10). Lipkin, D . , Divis, G. J., and Jordan, R. W., 155th National Meetings of the American Chemical Society, San Francisco, California, Preprints, Div. Petrol. Chem. (1968), 13, D61. (11). Mazoleyrat, J . P . , and Welvart, Z., Chem. Comm., (1972), 547. (12). Garst, J. F., and Barton, F. E., J. Am. Chem. Soc., (1974), 96, 523. (13). (a). Russell, G. A. and Lamson, D. W., J. Am. Chem. Soc., (1969), 91, 3967. (b). Fischer, H . , J. Phys. Chem., (1969), 73, 3834. (14). Lawler, R. G., in Lepley, A. R., and Closs, G. L. "Chemically Induced Magnetic Polarization", Wiley, New York, 1973. (15). (a). Daney, M . , Lapoyade, R., Mary, Μ., and Bouas-Laurent, Η., J . Organomental Chem., (1976), 92, 267. (b). Harvey, P. G., and Cho, H . , J. Am. Chem. Soc., (1974), 96, 2434. (c). Zieger, Η. E. and Gelbaum, L. T., J. Org. Chem., (1972), 37, 1012. (d). Panek, E. J., J . Am. Chem. Soc., (1974), 96, 7959. (16). Fu, P. P . , Harvey, R. G., Paschal, J . W., and Rabideau, R.W., J . Am. Chem. Soc., (1975), 97, 1145. (17). Electron transfer from the anthracene radical anion to The dihydroanthracenyl radical proceeds readily and com pletely. Bank, S. and Bockrath, B . , J. Am. Chem. Soc., (1972), 94, 6076, and references cited therein. (l8). Breslow, R., and Mazur, S . , J. Am. Chem. Soc., (1973), 95, 584. (19). House, H. O. and Weeks, P . D . , J. Am. Chem. Soc., (1975), 97, 2785. (20). Lowry, T. H. and Richardson, K. S . , "Mechanism and Theory in Organic Chemistry", pp. 146-147, Harper and Row, New York, 1976, and references cited therein. Downloaded by UNIV OF CALIFORNIA SAN DIEGO on February 5, 2016 | http://pubs.acs.org Publication Date: June 1, 1978 | doi: 10.1021/bk-1978-0069.ch021

(c).



358

(21). (22). (23).


(24). (25). (26). (27). (28).

Kochi, J. K. and Powers, J . W., J. Am. Chem. Soc., (1970), 92, 137. Mohammad, M. and Kosower, Ε. Μ., J. Am. Chem. Soc., (1971), 93, 2709. Mohammad, M. and Kosower, Ε. M . , J. Am. Chem. Soc., (1971), 93, 2713. Streitwieser, Α . , "Solvolytic Displacement Reactions", pp. 30-31, Mc Graw-Hill, New York, 1962, and references cited therein. Lowry, T. H. and Richardson, K. S . , Reference 20, p. 193. Bank, S . , Bank, J., Daney, M . , Labrande, B . , and BouasLaurent, H . , J . Org. Chem., (1977), 42, in press. Danen, W. C., Tipton, T . J. and Saunders, D. G., J. Am. Chem. Soc., (1971) 93, 5186. (a). Bank, S . , and Bockrath, B . , J. Am. Chem. Soc., (1971), 93, 430. (b). Bockrath, B. and Dorfman, L. M . , J. Am. Chem. Soc., (1974), 96, 5708.

RECEIVED D e c e m b e r 2 3 , 1 9 7 7 .


Radical, Anion, Anion-Radical Reactions with Organic Halides

Recommend Documents